Demo

Watch KubeIntellect debug a cluster

A recorded kq session: a plain-English question in, root cause and a proposed fix out — with every write held for human approval. Then run the same thing on your own cluster in three commands.

kubeintellect — recorded session

Recording of a real session · pause, scrub or replay with the controls

Run it on your own cluster

KubeIntellect runs next to your cluster, with your own LLM key — nothing goes through us. No cluster to hand? The setup wizard can create a local Kind cluster so you can try it on a laptop.

Step 1 — Install the server and the kq CLI

$pip install kubeintellect

Python 3.12+. Installs the kq client too (also available on its own with pip install kube-q).

Step 2 — Run the setup wizard

$kubeintellect init

Asks for your OpenAI or Azure OpenAI key, uses your kubeconfig (or creates a local Kind cluster), generates your API key and configures kq. kubeintellect status shows the key later. Prefer Docker Compose or Helm? See the self-hosting quickstart.

Step 3 — Ask

$kq -q "why is my pod crashlooping?"

Or just run kq for an interactive session. Pointing at a server elsewhere: kq --url http://localhost:8000 --api-key YOUR_KEY.

> Why is my pod crashing?

> Show all failing deployments

> What's using the most memory?

> What changed in the last hour?

Read-only by default; any change pauses for your approval. More in the quick install guide and the architecture.